CCAR-P Governance, Safety, and Risk Management Practice Question
A software company's internal AI review board is defining escalation criteria for its Claude-powered support assistant. The board wants a rule that reliably routes the highest-consequence cases to human specialists rather than relying on the model's own confidence statements. Which escalation design best achieves this?
⚠ Common exam trap
The trap here is trusting the model's own confidence or ambiguity judgments as the routing trigger, when those signals are uncalibrated and can be high precisely when the answer is wrong.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Route cases to specialists based on objective risk signals such as account tier, regulatory keywords, and prior complaint history.
Reliable escalation must be driven by signals that exist independently of the model's self-assessment. Objective criteria such as account tier, regulatory keywords, and complaint history correlate directly with consequence and can be evaluated deterministically, guaranteeing that the cases the board cares most about reach a specialist before a reply is finalized.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Route cases to specialists based on objective risk signals such as account tier, regulatory keywords, and prior complaint history.
Why this is correct
Objective signals are observable before or independently of the model's output, so routing does not depend on the model judging its own reliability. High-value accounts, regulated topics, and repeat-complaint histories are exactly where errors carry the greatest consequence, making this a deterministic and auditable way to guarantee specialist review.
- ✗
Ask the assistant to flag any conversation it finds ambiguous and forward those to specialists for a second opinion.
Why it's wrong here
This again delegates the routing decision to the model's self-assessment. Ambiguity as perceived by the assistant correlates weakly with business consequence: a straightforward but high-stakes request about a regulated product may feel unambiguous to the model and never be flagged, so the highest-consequence cases can slip through.
- ✗
Instruct the assistant to state a confidence percentage with each answer and escalate whenever it reports below ninety percent.
Why it's wrong here
Self-reported confidence is generated by the same model that produced the answer, so it is not a calibrated probability and can be confidently wrong. A threshold applied to an uncalibrated number gives false assurance: high-consequence errors may still arrive with a high stated confidence, bypassing the very routing rule the board intends.
- ✗
Sample five percent of all conversations at random for specialist review after the assistant has already replied.
Why it's wrong here
Random post-hoc sampling is a quality-assurance technique, not an escalation control. Because selection is independent of consequence, the highest-risk cases are reviewed at the same low rate as routine ones, and the customer has already received the reply, so the specialist cannot prevent the harm the board is trying to avoid.
About these practice questions
This CCAR-P question is part of Courseiva's 262-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Anthropic exam blueprint
This CCAR-P practice question is part of Courseiva's free Anthropic certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the CCAR-P exam.